{
  "fetched_at": "2026-10-24T17:58:07Z",
  "source_url": "https://www.kaggle.com/competitions/dsci-09/data",
  "pages": [
    {
      "name": "data-description",
      "content": "Offline\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p>A manager at the bank is disturbed with more and more customers leaving their credit card services. They would really appreciate if one could predict for them who is gonna get churned so they can proactively go to the customer to provide them better services and turn customers' decisions in the opposite direction.</p>\n\n<p>You can make 20 submissions per day. Once you submit your results, you will get a roc-auc score computed based on 40% of the test data. This score will position you somewhere on the leaderboard. Once the competition ends, you will see the final accuracy computed based on 100% of the test data.</p>\n\n<h2>Acknowledgements</h2>\n<p>Kaggle team for the dataset</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>The evaluation metric for this competition is <a href=\"https://towardsdatascience.com/understanding-auc-roc-curve-68b2303cc9c5\"> AUC Score</a>. The ROC curve is created by plotting the true positive rate (TPR) against the false positive rate (FPR) at various threshold settings. The true-positive rate is also known as sensitivity, recall or probability of detection[7] in machine learning. The false-positive rate is also known as probability of false alarm[7] and can be calculated as (1 − specificity). It can also be thought of as a plot of the power as a function of the Type I Error of the decision rule (when the performance is calculated from just a sample of the population, it can be thought of as estimators of these quantities). <a href=\"https://en.wikipedia.org/wiki/Receiver_operating_characteristic\"> WIKI</a></p>\n\n<h2>Submission Format</h2>\n<p><strong>For every customer (clientnum) in the dataset</strong>, submission files should contain two columns: CLIENTNUM and Attrition_Flag. Attrition_Flag should be a probability between 0 and 1.\n</p><p>The file should contain a header and have the following format:</p>\n<pre>CLIENTNUM,Attrition_Flag<br>752648883,0.5<br></pre>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - No AutoML\n - No Collaboration\n - Google is your friend!\n - Have fun!\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}